The acquisition and manipulation of genes is at the heart of biological research. Fast, accurate gene synthesis can reduce the cost and accelerate the speed cf research by allowing scientists to focus on experiments rather than spending time "cutting and pasting'' DNA. In addition, it has the potential to enable novel approaches to generating new biological functions by providing the raw material for "directed-evolution" techniques. Currently one of the main barriers to low-cost gene synthesis is the frequent introduction of sequence errors during the course of chemical synthesis. The aim of this research is to develop and implement novel techniques that will remove sequence errors from synthetic DNA and increase the fidelity of the gone synthesis process. If successful, this project will significantly lower the cost of gene synthesis, reduce delivery times, and increase the reliability of the process. In turn, these improvements will lead to more widespread use of gene synthesis in commercial and NIH-supported research and measurable increases in research productivity. We plan to commercialize the process through our existing gene synthesis business. Error reduction technology will allow shorter delivery time and lower costs for gene synthesis and enable the production of high quality libraries of gone variants. In addition, a number of the technologies we generate may lead to new products for mutation detection, for the production of very high quality oligonucleotides or for high fidelity PCR cloning.